Methodology

1. What a TVSM score is — and is not

A TVSM score is a risk tier, not a grade. It measures the gap between what a financial counterparty promises and what it reliably delivers, and the direction that gap is moving — trader exposure, not firm character.

For prop firms (TVSM-PF) the question it answers is concrete: “Will this firm pay me what I earned, and still exist to do it?” Every TVSM-PF variable traces back to that question.

TVSM 2.0 ships three frameworks over a shared spine: TVSM-PF (prop firms, 17 variables), TVSM-PL (platforms, 5 dimensions — documented at /methodology/platforms), and TVSM-BR (brokerages, 5 dimensions). This page documents TVSM-PF.

2. The scoring formula

Each of the 17 TVSM-PF variables is scored on a 1–10 integer scale. A firm scores at the highest level whose condition bundle is fully satisfied (AND-logic). Ladders are monotonic and single-axis: every rung scores the same observable; only the threshold moves.

TVSM-PF weights sum to 100, so the formula produces a 0–100 score directly. The trajectory adjustment is a deterministic event-delta engine — the severity-weighted sum of trailing-12-month event changes, normalized and clamped to ±3. Every published score surfaces its top 3–4 score-limiting predicates as named reason codes.

What we publish, and what we don’t. The criteria are public: every variable, every weight, the band boundaries, the cap classes, the evidence rules, and — on each firm page — the per-variable evidence trail with source links and verbatim extracts. What we deliberately do not publish are the exact rung thresholds inside each 1–10 ladder, cap trigger levels, and our detection heuristics. We are transparent about how we judge; we do not hand firms a manual for engineering their documents to sit exactly above a known trip-wire. Every published claim about a specific firm remains fully auditable through its cited sources.

3. Score bands

TVSM 2.0 uses risk-framed band labels. Boundaries are unchanged from v1.0.2.

Affiliate-link rule.Firms scoring below 50 receive no affiliate links. They carry a “Research Only” label. Public, permanent, non-negotiable. See the affiliate disclosure for the full firewall.

4. The six dimensions

Dimension weights sum to 100%. No display normalization required. Market profiles (Futures / Forex / Equities / Multi-asset) may swap and reweight variables within this framework — see the source spec for profile detail.

5. All 17 variables

Each variable scores 1–10 on a monotonic predicate ladder. Hard caps named here override the computed composite when triggered — see section 6 for cap mechanics.

6. Hard caps

Hard caps override the computed composite. Every applicable cap is evaluated; the lowest ceiling applies. Every cap applied must be visible on the firm page with which cap triggered, what the computed score would have been without it, and the specific evidence.

7. Evidence requirements

Every variable score must trace to a citation — preferably a primary source: the firm’s terms of service, official FAQ, a regulator filing, a verified payout proof, or a community thread with primary evidence (screenshots of dashboards, payout statements, email headers).

TVSM 2.0 uses a five-state evidence model: verified, limited, unknown, contradictory, and verification_pending. A variable that cannot be evidenced at the higher rungs of its ladder defaults to the catch-all level 1. Each citation includes the source URL, a verbatim source extract, and a verification date.

Primary regulatory sources. License and regulatory-status variables are verified against the public registers below. Individual firm pages cite the specific records used; these are the canonical registers the framework checks.

• NFA BASIC — US futures registration & disciplinary history
• CFTC — US derivatives regulator
• FCA Register — UK authorised-firm register
• SEC — US securities regulator & EDGAR filings

8. Freshness, trajectory, and effective pricing

Freshness and decay.Every variable has a freshness window specific to its dimension — payout-related variables decay fastest (a stale payout claim is a high-risk claim), regulatory and ownership variables decay more slowly. A variable that hasn’t been re-verified within its window drops to a lower confidence state, which can shift its score. The firm page shows the last verification date so the reader can see how fresh each spec is.

Trajectory. The trajectory adjustment captures direction, not level. It is a deterministic event-delta engine: each typed event (rule change, payout incident, ToS update, fee revision, payout-cap shift) is assigned a severity weight; the trailing-12-month sum is normalized and clamped to ±3. A firm with a strong base score that is trending sharply down can still publish at a score 3 points below its base — and the trajectory glyph (▲▲ / ▲ / ▬ / ▼ / ▼▼) is visible inline with the score so the direction is never hidden.

Dynamic effective pricing.The cost-and-value dimension scores the effective fee paid, not the headline price. Discount stacks, refundable resets, account-size optionality, and force-close-after-N constraints all feed into the effective fee ratio — so a 50%-off coupon stack that requires a re-evaluation after three payouts isn’t allowed to mask the true cost of access.

9. Independence firewall

Affiliate relationships never affect scores. Scores cannot be bought or influenced: no commercial relationship — including a scored firm’s own — can move a score. Firms we don’t editorially recommend do not receive affiliate links. This is stated publicly, in writing, and applies regardless of what commission rate is offered.

A 31/100 score is as important to publish as an 89/100 score. The negative scores are the trust signal that makes the positive scores meaningful. A site that only rates firms well is not a review site — it is an advertisement.

“Verified by firm” badge. Some profiles carry a small Verified by firmbadge in the header. It means the firm’s operator has reviewed the published facts on that page — specs, fees, payout terms, policies — and confirmed they are accurate. It is a provenance signal about the facts, not an endorsement of the score: a firm cannot move its own TVSM number, and the badge has no weight in the formula. A firm with a Caution-band score can be fact-verified; a firm we recommend may carry no badge at all. The two are independent by design.

10. Version history

Versioning is semantic and per-framework. Patch bumps (2.0 → 2.0.1) are predicate-rubric clarifications — no weight changes, no variable additions, no new caps. Minor bumps add variables or change evidence standards. Major bumps change dimension weights by > 3% in absolute terms or add a framework. On any version bump, every scored firm within that framework is fully recomputed and a score_history snapshot is recorded with the methodology-driven delta.

11. Standards & references

TVSM is our own framework, not a reproduction of any external standard — but its risk-tier orientation (measuring the gap between what a financial counterparty promises and what it reliably delivers) is deliberately consistent with established investor-protection principles. The conceptual grounding draws on, and can be cross-read against, the following independent standards bodies:

• IOSCO — the International Organization of Securities Commissions’ Objectives and Principles of Securities Regulation, the global baseline for the investor-protection and counterparty-disclosure concepts the Payout Integrity and Transparency dimensions test.
• CFA Institute — the Code of Ethics and Standards of Professional Conduct, whose fair-dealing and full-and-fair-disclosure principles inform how we score pricing honesty and rule-change communication.
• ESMA Investor Corner — the European Securities and Markets Authority’s guidance on retail leveraged-product risk, relevant context for the leverage and drawdown variables.

These are corroborating standards, not the source of any individual score. Every score is computed from the firm’s own primary documents, cited inline on each firm page.